BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:Towards Domain-Specific Data Management
DTSTART:20120607T150000
DTEND:20120607T160000
DTSTAMP:20260407T230252Z
UID:fb4033151d0fef110bcf4c733c840a285426ea2eeda5e0512413ba61
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Yanif Ahmad\, Johns Hopkins University\, Baltimore\, USA
 \nAbstract:\nTo overcome scaling challenges\, today's computing applicatio
 ns are increasingly exploiting domain-specific properties in their data\, 
 computation\, and infrastructure models. In data management\, specializati
 on of monolithic relational database management systems that cannot servic
 e the needs of evermore diverse datasets has led to many new classes of da
 ta management systems. In programming languages\, embedded domain-specific
  languages have emerged as a popular technique to express specialized prim
 itives\, operators and optimizing toolchains. We argue that data managemen
 t systems should expose facilities for domain-specific data models\, where
  we can leverage known mathematical properties and representations of our 
 data. Combining mathematical modeling with declarative querying facilitate
 s rich exploratory access to data\, and compact data representations with 
 tunable approximation.\nIn this talk\, I will first present Pulse\, a quer
 y processor for continuous-time databases based on temporal polynomial mod
 els. Pulse uses piecewise polynomials to provide a compact\, approximate r
 epresentation of the input dataset and processes queries by solving simult
 aneous equation systems in contrast to set-at-a-time record processing. Pu
 lse is able to achieve significant performance improvements by directly pr
 ocessing polynomials prior to discretization\, and by exploiting user-defi
 ned precision bounds to reduce computation overheads. Beyond polynomial mo
 dels\, I will also discuss our ongoing work with processing queries on sta
 tistical models\, specifically probabilistic graphical models\, applying j
 oint incremental query processing and inference techniques in the BLOG (Ba
 yesian Logic) programming language.\n\nBio:\nYanif Ahmad is an Assistant P
 rofessor in the Department of Computer Science at the Johns Hopkins Univer
 sity. His research goals are to enable insightful monitoring of large stre
 aming datasets\, and to facilitate easier declarative construction of scal
 able systems in novel computing applications. In addition to the talk topi
 cs\, Yanif's ongoing work includes exploring joint database and compiler-s
 tyle optimizations for large-scale data processing and analytics\, and pro
 tein data management for a petabyte-scale molecular and drug design datase
 t in collaboration with the Johns Hopkins Medical School. He received his 
 PhD from Brown University\, and has been a postdoctoral associate with the
  Database Group at Cornell University.
LOCATION:BC 01 https://plan.epfl.ch/?room==BC%2001
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR